A Diet Abroad: Food Scanner

No more fear of dietary restrictions.

Team/Roles: Stuart Giles: UI, wireframes, research, presentation
Laura Hunter: Copywriting, persona development, project management
Riyaz Javanmardi: Industrial design, materials design

Date: October - December, 2014

Skills used: Wireframing, slide design, presentation, interviewing, usability testing, survey design

Project Summary

     As someone with an anaphylactic allergy to tree nuts, I must be acutely aware of what's in what I'm eating at all times: a single mistake will mean a hospital visit or, if I don't have an EpiPen with me, possibly even my death. This is especially a hassle when I'm traveling and don't necessarily know the language, and can be a significant source of stress. This topic was on my mind during my User-Centered Design class, so when we were given free rein to decide the topic of our quarter-long projects, my group decided that this was a hole that design could fill. Food restrictions don't just affect people like me with severe allergies: there are also less severe but still unpleasant cases, like Celiac Disease, and even religious concerns such as keeping Kosher or Halal which could find benefit in a solution for the same problem.

Ideation

     Our initial idea for the product was to create a phone and web app which crowdsourced information on restaurant and other food seller "allergy friendliness," including metrics such availability of dish ingredients, willingness to make substitutions, etc. In startup parlance, "like Yelp for allergies." We realized, however, that there were a number of flaws with this idea: even discounting the difficulty of crowdsourcing so much information, many people wouldn't have data on their phones to check it when out of their home countries! We hit a snag, then, until we came upon the a technology called Raman spectroscopy, which allows for the scanning and identification of materials using projected light.

     The ability to know the chemical makeup of a food item was absolutely relevant to our project, so we brainstormed: what do people with serious allergies already carry around everywhere with them? An epinephrine auto-injector! The idea took form from that: what if we could design an auto-injector that not only reduced the stigma of carrying around a medical device everywhere by looking like a regular tech toy, but could also scan food and discreetly alert the user to the presence of problem ingredients?

Research

     Initial research involved first designing a survey on food habits and distributing it to both general audiences and audiences with dietary restrictions (such as online communities like Reddit's Food Allergies and Gluten Free groups) to pinpoint differences in habits between people with and without dietary restrictions. Following that, I conducted interviews with several people with moderate to severe dietary restrictions, and asked each of them to keep a diary of their food-related decisions for two weeks. While I was busy with that, Riyaz designed several different ideas for the form factor of the device, 3D printing each, and we met to discuss them each time he iterated until we came to the final form factor.

Refinement and Usability Testing

     I designed and wireframed a UI for the phone app based on the needs expressed in the survey, interview, and diary results. This app, designed as a companion app for the hardware, not only included scanner features such as the ability to create a dietary restriction profile, but also a number of companion features, such as restaurant menus with allergen warnings, and Yelp review integration. Importantly, it also allowed users to download any information provided, so they could peruse it offline in a place they may not necessarily receive connection.

     Finally, we constructed a click-through prototype of the app using POP, and used that to conduct usability tests with the interview and diary participants, using a think-aloud protocol. Insights gained from these tests were used in the final presentation and the construction of the design spec. Unfortunately, time constraints were too tight to construct a high-fidelity prototype in time, so the wireframes were used in the spec. This project was an excellent look at how to proceed through the user-centered design process, and I feel fortunate to have been in a group with an experienced program manager and industrial designer to learn from.

Design Spec